Heat of Solution of Boron Trifluoride in Water and Aqueous

Heat of Solution of Boron Trifluoride in Water and Aqueous Hydrofluoric Acid. D. A. Scarpiello, W. J. Cooper. J. Chem. Eng. Data , 1964, 9 (3), pp 364...
0 downloads 0 Views 227KB Size
Heat of Solution ob Boron Trifluoride in Water and Aqueous Hydrofluoric Acid D. A. SCARPIELLO and W. J. COOPER Callery Chemical Co.,Callery, Pa. The heat of solution of boron trifluoride in water and in dilute aqueous hydrofluoric acid was measured in a rotating bomb calorimeter. The heat of solution, AHm, was -24.7 f 0.5 kcal. mole-' for both cases over the concentration range from 0.48 to 1.17 molal in boron trifluoride. The calculated heat of formation of BFa(0H) - , a q , l is -363.1 kcal. mole-'.

RATHER LIMITED

experimental data are available in the literature for the heat of solution of boron trifluoride in water. Hammer1 ( I ) and Laubengayer and Finlay ( 3 ) report values which indicate the heat of solution to be independent of the final concentration over the range from 0.2 to 3 molal in boron trifluoride. We have obtained heat of solution values in reasonable agreement with the literature and have confirmed the absence of concentration dependence over the more limited range of 0.55 to 1.17 molal. The heat of solution has also been found to be unaffected by low concentrations of hydrofluoric acid over the range of 0.48 to 1.01 molal in boron trifluoride. EXPERIMENTAL Calorimeter. The rotating-bomb calorimeter was constructed from Argonne National Laboratory design CE3986-A which is similar to that described by Hubbard (2). The bomb was made entirely of "A" nickel and had an internal volume of 350 cc. Temperature measurements were made with a calorimetric platinum resistance thermometer in conunction with a G-2 Mueller bridge and high-sensitivity galvanometer. Calibration. The calorimeter was calibrated by combustion of standard benzoic acid, NBS 39h. Eight combustion experiments gave a mean energy equivalent for the standard calorimeter of 35,454 cal. ohm ' with a standard deviation of the mean of *34 cal. ohin-'. All energies are expressed in terms of the defined calorie, equal to 4.184 (exactly) joules. Materials. Boron trifluoride (Allied Chemical) was passed through a mercury bubbler into a high-vacuum system where non-condensables were removed. Infrared spectra showed no lines other than those for boron trifluoride. Hydrofluoric acid solutions were made up by dilution of ACS reagent 4 8 5 H F (Allied Chemical). Procedures. Thin-walled Pyrex bulbs were filled with water, degassed, and sealed under vacuum. The bulbs, containing a weighed amount of water, were placed in the bomb. The bomb was evacuated and then pressurized with the desired amount of boron trifluoride, in the range from 0.9 to 1.9 atmospheres. After the usual equilibration period in the calorimeter, rotation of the bomb was started, breaking the glass bulb and initiating the solution reaction. Rotation of the bomb was continued throughout the remainder of the run. Product Analyses. Completeness of solution was checked by infrared analysis of the vapor after solution, with no boron trifluoride being observed. Total boron and fluorine in solution were determined by decomposition of the fluoboric acids in the presence of calcium chloride. The FIB mole ratio obtained from the analyses was 3.00 i 0.02. The boric acid was titrated in the presence of mannitol and the fluorine determined as the calcium fluoride pre-

364

cipitate. A small amount of nickel appeared in the solutions from the side reaction with the bomb surfaces. The amount varied from 2-4 mg., determined as the dimethylglyoxime complex. The nickel was removed by precipitation as the hydroxide before titration of the boric acid. EXPERIMENTAL RESULTS The experimental results of the solution of boron trifluoride in water are summarized in Table I. A correction was applied for the energy of vaporization of the water vaporized when the bulb was broken. Over the concentration range from 0.55 to 1.17 molal, there was no significant dependence of concentration. The mean value of A E from Table I gave, on conversion t o constant pressure, a standard heat of solution, AHm8, of -24.7 f 0.4 kcal. mole-' BF3-'. The uncertainty interval was taken as twice the standard deviation of the mean. The experimental data for the solutions of boron trifluoride in 0.32 molal hydrofluoric acid are summarized in Table 11. The mean heat of solution, AH*%,and uncertainty 0.6. Again, there was no observed interval are -24.7 concentration dependence. I n all of the experiments, correction for the nickel reaction was assumed t o be negligible.

Table I. Boron Trifluoride-Water Solution Data Exp. No. 1 2-

3 4 6 8 9 10 11

Mass H20 (Grams)

Mmoles BFs

17.92 31.10 27.52 20.90 23.91 26.08 25.00 16.96 16.07

13.1 27.3 24.1 17.9 14.6 14.3 14.7 18.8 18.8

~~

~

-~

~

Final BF? -AE soh. Molality (Kcal. Mole 0.731 0.878 0.876 0.856 0.611 0.548 0.588 1.108 1.170

Mean Std. Dev. of Mean

')

23.9 23.2 24.4 24.8 24.4 24.1 23.5 24.7 24.0 24.1 rk0.2

Table II. Boron Trifluoride-Hydrofluoric Acid Solution Data

Exp. No. 3 4 5 6 7 9

Mass HF Soln. (Grams)

Mmoles BF,

22.15 19.09 20.10 18.97 19.41 19.41

22.3 19.3 15.9 13.9 10.4 9.4

Final BF7 -AE soln. Molality (Kcal. Mole-') 1.007 1.001 0.791 0.733 0.536 0.484

Mean Std. Dev. of Mean

23.4 25.2 24.2 24.1 24.1 23.6 24.1 +0.3

JOURNAL OF CHEMICAL AND ENGINEERING DATA

DISCUSSION

T h e heat of solution of boron trifluoride in water observed in this work is slightly lower than that obtained by Hammer1 (1) and Laubengayer and Finlay ( 3 ) . Their results adjusted to 25" C. are -25.7 and -25.1 kcal. mole-', respectively. Wamser (6) and Ryss and his coworkers ( 4 ) have studied the hydrolysis equilibria of BF3 solutions. Wamser reports the initial rapid reaction of BF3 with water to be: B F B +~ (n + 1) HzO = HBFSOH.nH20

~

(1)

with further hydrolysis being much slower. Ryss and Elkenbard ( 4 ) have measured the heat of solution of KBFsOH in water and aqueous NaOH. From these data, the heat of formation of BF30H-, has been derived as -363.1 kcal. mole-'. Assuming the solution reaction under the conditions of our study to be represented by Reaction 1, the heat of formation of BFsOH-, may be derived as -363.1 kcal. mole-', using -270.1 kcal. mole-' for the heat of formation of BF,e (7). T h e exact agreement with Ryss is undoubtedly fortuitous. Our solution data in dilute H F indicate that, a t low HF/BF3 ratios (0.3-0.6), the heat of solution is unaffected by the presence of H F and the predominant species may still be assumed to be BFsOH-. Wamser ( 5 ) has shown that higher H F / B F j ratios (> 3) drive the equilibrium toward formation of BF,- as the major species. We have measured a single heat of solution a t an HF/BFs ratio of approximately 3. The observed heat of solution of -27.0 kcal. mole is in the direction one would expect, since the heat of formation of BF, o4 of -373.7, derived from the data of Ryss ( 4 ) , gives a calculated heat for the reaction: BF,,

+ H F .n H 2 0(aq 1 = HBF,. nH?O(aq

(2)

of -27.9 kcal. mole-'.

The use of Equation 1 to represent the solution reaction in water is an oversimplification, since all of the possible acid species are present to some extent. This does not prevent, however, the discussion of the heat effects on solution as though they were produced only by the predominant species. HBF30H. Wamser (6) presents a more detailed hydrolysis scheme, but it is apparent from the results of the heat of solution measurements that the over-all heat balance of the complex equilibria is such that the measured heat is not significantly affected by minor changes in concentration of the various species. ACKNOWLEDGMENT

We wish to thank W.E. Hazen for assistance in performing the analyses. LITERATURE CITED Hammerl, H., Compt. rend. 90, 312 (1880). Hubbard, W.N.. Katz. C.. Waddineton, G.. J . Phvs. Chem. 58, 142 (1954). Laubengayer, A.W., Finlay, G.R., J . A m . Chem. Soc 65, 884 (1943). Ryss, I.G., Slutskaya, M., Compt. rend. acad. sei. URSS 52, 417 (1946). Ryss, I.G., Slutskaya, M., Dohlady Akad. Nauk SSSR57, 689 (1947). Ryss, I.G., Elkenbard, A.G., Ibid. 91, 865 (1953). Ryss, I.G., Slutskaya, M., Palevskaya, S.D., Zhur. Fiz. Khim. 22, 1322 (1948). Wamser, C.A., J . A m . Chem. SOC.70,2109 (1948). Ibid. 73, 409 (1951). Wise, S.S., Margrave, J.L., Feder, H.M., Hubbard, W.N., J . Phys. Chem. 65, 2157 (1961). I

RECEIVED for review November 7 , 1963. Accepted March 12, 1964. This work was supported by the Air Force Office of Scientific Research under Contract AF-49(638)-1052.

Equilibrium Diagram of the System Nb205-WO3 LEON J. FIEGEL, GANESH P. MOHANTY', and JAMES H. HEALY' A. 0. Smith Corp., Milwaukee, Wis. An equilibrium diagram of the Nb205-WO1 system has been presented. The existence of two new compounds, Nb205-WO3 and Nb20je3W03, has been postulated. The work of Kovba and Trunov on Nb205.3WOa ( 0 , b = 12.190A, c = 3.934A.) has been verified. Experimental procedures included room temperature x-ray diffraction of previously annealed samples, high temperature x-ray diffraction, and differential thermal analysis. Some work was also done on the low temperature transformation of WOs and W03-Nb205 compounds.

AS P A R T of a project concerning the oxides of the refractory metals, the phase equilibrium of the system Nb205-W03was investigated. The one published report o n this system was done by GoldSchmidt ( I ) . His results indicated that WO1 was soluble in N b 2 0 5to the extent of the existence Of a 6o % wo'' He compound existing a t the 25% Nb205-75'"c WOi composition. Present address: Florida State University, Tallahassee, Fla. address: Spindletop Research Institute, Lexington, Ky.

* Present

VOL. 9, No. 3, JULY 1964

EXPERIMENTAL Sample Preparation. The oxide samples used in this investigation were prepared from materials obtained from A.D. Mackay and Co., Inc. The purities were 99.9+% for the Nb205 and 99.995 for the WO?. Impurities detected spectroscopically in these materials were Si, Mg, Co, Zr, and Ni for Nb,Oj and Si, Al, and Mg for WO?. Before mixing, the powders were prefired in air for 24 hours at 800"C. During this treatment and all subsequent annealing treatments, the samples were enclosed in platinum crucibles

365